As the U.S. HIV epidemic continues to unfold, injection drug use accounts for an ever increasing proportion of HIV infections. Current prevention efforts have failed to adequately slow the spread of HIV among injection drug users, Community interventions have been successful in promoting a wide variety of health behaviors but are only now beginning to be tried in efforts to prevent HIV. We propose to evaluate a multifocal community intervention to slow the spread of HIV among out-of-treatment injection drug users. The project combines the principles of community ethnography to identify and gain access to social networks of drug users, and diffusion theory to mobilize drug user's natural social networks around community norms for safer injection and safer sex. We will: . Evaluate the intervention's impact upon community norms for safer injection drug use and safer sex. . Examine the intervention's impact in promoting safer injection and safer sex. . Assess relationships between exposure to individual intervention components and safer behaviors. . Assess the extent to which peer support and peer communication mediate the effect of the intervention on study outcomes. . Estimate the effectiveness and cost-effectiveness of the intervention in averting new HIV infections. The evaluation of the intervention entails surveys of injection drug users conducted continuously in the intervention community (Sacramento) and comparison community (San Diego) before, during, and after implementation of the intervention in Sacramento. One thousand (1,000) survey respondents (500 each in Sacramento an San Diego) each year will be interviewed annually in years 1 through 4 of the study. The intervention, implemented in years 2 and 3, is designed to reach all injecting drug users in the community. The intervention consists of six interlocking components that can be implemented by a five person outreach team. The components include; 1) social network-targeted outreach; 2) a grassroots social network intervention designed to build community norms and peer support for risk reduction; 3) a high volume mobile needle-exchange; 4) an advocacy group organized by and for injection drug users; 5) specialized media targeted to drug users; 6) and a community advisory board composed of representatives of community agencies serving injection drug users to assure consistent dissemination of prevention messages. Data analysis will compare trends in community norms for safer injection and safe sex and preventive behaviors across cities to assess the effectiveness of the intervention.